Contraception as a pathway to better child nutrition and health

India is home to one-third of the world’s stunted children, and half of all under-five mortality can be attributed to undernutrition. Existing literature has shown an association between larger families and poorer child health outcomes. Based on analysis of data from the National Family Health Survey, 2019-2021, this article finds that reducing fertility through increased use of contraception can lead to significant improvements in child health and malnutrition indicators.

Child malnutrition is a persistent and ever-increasing public health concern across the underdeveloped and developing world. Per the World Health Organization (WHO) 2021 estimates, 149 million children under the age of five years are stunted, and 45 million are wasted.¹ Apart from the physiological impacts, child malnutrition poses huge economic costs, ranging from 2-3% of a country’s GDP (gross domestic product) to as high as 17% in the most severely impacted regions (Dione et al. 2013, World Bank, 2006). India has particularly been struggling to tackle child malnutrition. Despite decades of investments in child health, India continues to be home to one-third of the world’s stunted children, and half of all under-five mortality can be attributed to undernutrition.

Child malnutrition and its various determinants have been extensively studied, with evidence suggesting that inadequate dietary intake, unhygienic household environment, improper sanitation, and lack of proper breastfeeding are potential causes of undernutrition (Chandrasekhar et al. 2017, Ijarotimi 2013, Reading 2008). Additionally, Imai et al. (2014) and Smith et al. (2016, 2000) have established that maternal education, health insurance and empowerment can substantially impact child health indicators.

In recent research (Mookerjee, Ojha and Roy 2023), we make a novel attempt to establish the causal impact of contraception use on child health and explain plausible mechanisms for this relationship. Benefits of family planning and access to contraception are widely recognised, ranging from reduction in the probability of high-risk or unintended pregnancies, improvements in maternal health, higher child survival rates, and reduced risk of premature births or low birthweights through improved birth spacing (Conde-Agudelo, Rosas-Bermúdez and Kafury-Goeta 2006, Rau, Sarzosa and Urzúa 2021, Miller and Karra 2020, Molitoris 2017). The associations between poorer child health outcomes and larger families have been well-documented in the literature (Silles 2010, Knodel, Havanon and Sittitrai 1990, Horton 1986, Blake 1981). Most of these studies rely on the resource dilution hypothesis or the quantity-quality trade-off, suggesting that parental investments, time and care per child become scarce as families become larger, in turn leading to worse outcomes for children (Becker and Tomes 1976, Becker and Lewis 1973, Siegers 1987).

Ironically, although India was the first country in the world to launch a national programme for family planning in 1952, according to the National Family Health Survey (NFHS)-5 (2019-2021), stunting in children under five has increased in 13 out of 22 states since 2015-16 and wasting has increased in 12 states. This raises troubling questions about the efforts to address such chronic health issues and leads us to explore further the channels that may explain malnutrition in India to provide new insights for tackling it better.

Our study

The outcomes of interest in our study are the standardised anthropometric measures of children under five years, namely, their height-for-age (HAZ) and weight-for-age (WAZ) z-scores, which reflect their nutrition and growth status in the short and long term. These are usually reported as standard deviations (SD)² above or below the international reference population. We also explore the likelihood of a child being stunted or underweight. Our main explanatory variable is a binary indicator of whether a couple has ever used any contraceptive measures to delay pregnancy.

The primary econometric challenge in estimating this impact is that a couple’s contraception use can be ‘endogenous’ for various reasons. There may be characteristics of the couple such as education, that cause them to simultaneously use contraception and focus more on children’s nutrition – making it difficult to claim that use of contraception is causing children’s improved nutrition. Further, there may be ‘reverse causality’ if the poor health or nutritional status of older kids makes couples more likely to use contraception for future pregnancies to ease the financial burden of having more children. Finally, there can be measurement errors due to misreporting.

To alleviate such concerns, we exploit exogenous variation in the average district-level exposure of women to family planning messages through mass media including television, radio and newspaper, as an ‘instrument’ to proxy the couple’s contraception use.³

Key findings

Our study reveals that the use of contraception by a couple, leads to a 1.4 SD increase in a child’s HAZ and a 0.47 SD increase in WAZ. We also find that contraception use reduces the probability of a child being stunted and underweight by 17.9 and 10.9 percentage points, respectively. Notably, we also find larger impacts on a girl child’s HAZ and WAZ, in poorer families, and those residing in rural areas.

Recognising that our results may be sensitive to some econometric challenges, we conduct a host of robustness tests. To address the concerns about the disproportionate representation of children in smaller versus larger families in the study sample, we first re-estimate our model using the average child characteristics in each household and find similar results. Secondly, there may be reason to believe that other district characteristics, such as number of health care centres, women’s access to prenatal and postnatal care, frequency of visits to Anganwadi (childcare) or ICDS (Integrated Child Development Services) centres etc. might correlate with one’s awareness and exposure to family planning and potentially invalidate our instrument. To alleviate such concerns, we also control for a set of district-level variables and continue to find similar significant impacts. Third, we re-do our analysis using data from NFHS-4 (2015-16) to alleviate any concerns about the Covid-19 pandemic confounding our results and we confirm that the problem is persistent – parental family planning practices have been a significant predictor of child health over a long period of time. Lastly, we conduct falsification tests by repeated estimations of the model where we randomly associate the malnutrition indicators of a child from one household to the contraception decisions of some other household to find no significant impacts, further validating our findings.

We also establish the quantity-quality trade-off as a plausible mechanism explaining the impact of contraception use on child malnutrition. We find that a couple’s contraception use statistically significantly reduces household size. Further, we demonstrate that an exogenous increase in the household size, instrumented by multiple/twin births, leads to a reduction in children’s anthropometric measures (HAZ/WAZ).

Concluding remarks

Despite various policies under the National Food Security Act (NFSA), 2013, India is slipping into a vicious cycle of malnutrition, and child health continues to be one of the most critical public health concerns for the country. In light of this, we explore an important factor explaining poor child health and find that parental use of contraception can have significant benefits in terms of better anthropometric measures of their children as well as reduced likelihood of being stunted or underweight.

Nevertheless, our findings should be interpreted with a few limitations in mind. More detailed information on exactly when the couple used contraception or which pregnancy the use of contraception has tried to affect/delay may have led to a richer analysis. We also believe that a child’s health indicators would also be affected by the father’s characteristics (such as age, education, health status, employment), which we were unable to incorporate since only a small subset of men were interviewed and limiting the study sample to them would lead to a loss in statistical power. Besides, we posit that lengthening of birth intervals can be explored as another potential mechanism explaining the relationship between contraception and child health, in future research.

Through this study, we emphasise the critical role of family planning measures in addressing child ill-health. Given that improved nutritional outcomes for children ultimately contribute to better human capital accumulation, enhanced productivity, and reduced healthcare costs, this has significant policy implications. While factors like sanitation, food security, and maternal healthcare have been widely studied as key determinants of child health, our research highlights the need for greater attention to family planning practices, particularly by promoting contraceptive use to improve child nutrition in a country grappling with high population levels and the burden of child malnutrition.

Notes:

Malnutrition among children can result in low height-for-age (stunting), low weight-for-height (wasting), and low weight-for-age (underweight).
Standard deviation is a measure used to quantify the amount of variation or dispersion of a set of values from the mean (average) value of that set.
Instrumental variables can be used in empirical analysis to address concerns about endogeneity. The instrument (here, exposure to family planning messages) is correlated with the explanatory variable (contraceptive use) but does not directly affect the outcome of interest (children’s nutritional status). The instrument can therefore be used to measure the causal relationship between the explanatory factor and the outcome of interest.